| Week | Date | Subject | Textbook | Reading* | Task & Due (All deadlines are FRIDAY at 5 PM) |
|---|---|---|---|---|---|
| 1 | 8/28, Thursday | Introduction: History and examples of machine-learning (ML) applications. ML experiments, Part 1: Design and resampling methods. | Ch. 1; Ch. 20.1-20.6 |
PC: Ch. 1 ISL: Ch. 5 |
|
| 2 | 9/4, Thursday | Supervised learning: Multiple classes. Regression. Model selection and generalization. | Ch. 2 | ISL: Ch. 2.1.3-2.1.5 ISL: Ch. 3.1-3.2 |
|
| 3 | 9/11, Thursday | Bayesian decision theory: Classification, risk, and discriminant functions. Naive Bayes. | Ch. 3 | PC: Ch. 2.1-2.3 | Start thinking/planning your course project and writing project proposal; Homework #1 posted |
| 4 | 9/18, Thursday | Parametric methods: Maximum likelihood. Bias and variance. Bayes estimator. Classification. Regression. Model selection. | Ch. 4 | PC: Ch. 3.1-3.5 | |
| 5 | 9/25, Thursday | Multivariate methods: Parameter estimation. Missing values. Multivariate normal. Classification and regression. | Ch. 5 | PC: Ch. 2.4-2.6 ISL: Ch. 4.3-4.4 |
Homework #1 due; Homework #2 posted |
| 6 | 10/2, Thursday | Dimensionality reduction: Principal component analysis (PCA). Embedding. Factor analysis. Singular value decomposition. Multidimensional scaling. Linear discriminant analysis. t-SNE. Applications. | Ch. 6 | PC: Ch. 3.7-3.8 ISL: Ch. 6; 12.2 |
Project Proposal due |
| 7 | 10/9, Thursday | Fall Break (no class) | |||
| 8 | 10/16, Thursday | Clustering: Mixture densities. k-means. EM algorithm. Spectral and hierarchical methods. ML experiments, Part 2: Evaluation: measuring and comparing performance. | Ch. 7; Ch. 20.7-20.15 |
PC: Ch. 3.9 ISL: Ch. 12.4 |
Homework #2 due; Homework #3 posted |
| 9 | 10/23, Thursday | Nonparametric methods: Density estimation and extension to multivariate data. Classification. Nearest neighbor. Regression. | Ch. 8 | PC: Ch. 4.1-4.6 | |
| 10 | 10/30, Thursday | Decision trees: Classification. Regression. Pruning. Multivariate trees. Linear discrimination: Generalizing the linear model. Geometry. Parametric discrimination. Logistic discrimination. Applications. | Chs. 9, 10 | PC: Ch. 5.1-5.4 ISL: Ch. 8.1 |
|
| 11 | 11/6, Thursday | Multilayer perceptrons: Introduction and background. Training. Deep Learning, Part 1: Multilayer. Backpropagation. Overtraining. Autoencoders. Fully connected neural networks (FCNNs). | Ch. 11 | PC: Ch. 6.1-6.4 ISL: Ch. 10.1-10.2 |
Homework #3 due; Homework #4 Posted |
| 12 | 11/13, Thursday | Deep Learning, Part 2: Multiple hidden layers. Regularization. Convolutional layers and image analysis. Learning sequences. Generative adversarial networks (GANs). Transformer models. Applications. | Ch. 12 | PC: Ch. 6.8; 6.10 ISL: Ch. 10.3; 10.7 |
Progress Report due |
| 13 | 11/20, Thursday | Kernel machines: Optimal hyperplanes. SVM. Kernel trick. Multiple and multiclass kernels. Regression. Kernel Dimensionality Reduction. | Ch. 14 | PC: Ch. 5.11 ISL: Ch. 9 |
Homework #4 due |
| 14 | 11/27, Thursday | Thanksgiving Break (no class) | |||
| 15 | 12/4, Thursday | Combining multiple learners: Generating diverse learners. Methods for combining. Voting. Bagging Boosting. Fine-tuning and cascading. Reinforcement learning. | Ch. 18 | PC: Ch. 9.5-9.7 ISL: Ch. 8.2 |
|
| 16 | 12/9, Tuesday | Presentation of term projects | |||
| 17 | 12/16, Tuesday | Final exam | 5:20 pm - 8:20 pm in PHIL 109 | ||
| 18 | 12/19, Friday | Term Project (final) Report Due (no class) | Due at 5 pm on the day. No late submission. | ||
| END | |||||
| 📌 The schedule is subject to changes and adjustments. | |||||
| Textbook: E. Alpaydin, Introduction to Machine Learning, 4th Edition. The MIT Press, 2020 |
|||||
| *Readings refer to the recommended references:
▪ ISL: G. James, D. Witten, T. Hastie, and R. Tibshirani, An Introduction to Statistical Learning: with Applications in Python. Springer, 2023 ▪ PC: R.O. Duda, P.E. Hart, and D. G. Stork, Pattern Classification, 2nd Edition. Wiley, 2001 |
|||||